Friday, 2016-07-08

« Thursday, 2016-07-07 Index Saturday, 2016-07-09 »
*** amotoki has joined #openstack-ironic00:00
*** hoangcx has quit IRC00:00
*** phuongnh has quit IRC00:01
*** wajdi has joined #openstack-ironic00:01
openstackgerritThiago Paiva Brito proposed openstack/ironic: Allow to enroll nodes with oneview driver.  https://review.openstack.org/32845100:04
*** amotoki has quit IRC00:04
*** chlong has quit IRC00:05
*** rama_y has quit IRC00:07
openstackgerritThiago Paiva Brito proposed openstack/ironic: Allow to enroll nodes with oneview driver.  https://review.openstack.org/32845100:09
openstackgerritThiago Paiva Brito proposed openstack/ironic: Allow to enroll nodes with oneview driver  https://review.openstack.org/32845100:09
*** rook has quit IRC00:10
*** david-lyle has joined #openstack-ironic00:14
*** PollyZ has joined #openstack-ironic00:35
*** joprovost has quit IRC00:39
*** PollyZ has quit IRC00:39
*** joprovost has joined #openstack-ironic00:39
*** amotoki has joined #openstack-ironic00:40
*** vishwanathj has quit IRC00:41
*** joprovost has quit IRC00:43
*** rajinir has quit IRC00:45
*** hoangcx has joined #openstack-ironic00:45
*** david-lyle has quit IRC00:46
*** Sukhdev has joined #openstack-ironic00:46
*** amotoki has quit IRC00:48
*** amotoki has joined #openstack-ironic00:52
*** garthb has quit IRC01:01
*** phuongnh has joined #openstack-ironic01:06
*** amotoki has quit IRC01:09
*** david-lyle has joined #openstack-ironic01:13
*** david-lyle has quit IRC01:18
*** dhellmann has quit IRC01:19
*** dhellmann has joined #openstack-ironic01:26
*** chlong has joined #openstack-ironic01:27
*** joprovost has joined #openstack-ironic01:28
*** dhellmann has quit IRC01:31
*** dhellmann has joined #openstack-ironic01:31
*** zzzeek has quit IRC01:34
*** watanabe_isao has quit IRC01:36
*** dhellmann has quit IRC01:39
*** dhellmann has joined #openstack-ironic01:39
*** joprovost1 has joined #openstack-ironic01:39
*** joprovost has quit IRC01:40
*** joprovost1 is now known as joprovost01:40
*** aNupoisc has quit IRC01:40
*** dhellmann has quit IRC01:45
*** dhellmann has joined #openstack-ironic01:46
openstackgerritTan Lin proposed openstack/ironic: Add a doc about appending kernel parameters to boot instances  https://review.openstack.org/32084801:47
*** dhellmann has quit IRC01:51
*** dhellmann has joined #openstack-ironic01:52
*** dhellmann has quit IRC01:59
*** dhellmann has joined #openstack-ironic02:00
*** amotoki has joined #openstack-ironic02:01
*** chlong has quit IRC02:08
openstackgerritTan Lin proposed openstack/ironic: Add a new DB API get_missing_conductors()  https://review.openstack.org/32502602:11
*** Goneri has joined #openstack-ironic02:21
*** zzzeek has joined #openstack-ironic02:30
*** zzzeek has quit IRC02:31
*** zzzeek has joined #openstack-ironic02:32
*** tiendc has joined #openstack-ironic02:32
*** rloo has quit IRC02:34
*** yonglihe has joined #openstack-ironic02:38
*** amotoki has quit IRC02:39
*** ametts has joined #openstack-ironic02:40
*** Sukhdev has quit IRC02:45
*** amotoki has joined #openstack-ironic02:48
*** ChubYann has quit IRC02:56
*** joprovost has quit IRC02:59
*** amotoki has quit IRC03:00
*** yuanying has quit IRC03:04
*** sdake has joined #openstack-ironic03:08
*** ChubYann has joined #openstack-ironic03:09
*** sdake_ has joined #openstack-ironic03:10
*** amotoki has joined #openstack-ironic03:12
*** sdake has quit IRC03:14
*** jaybeale has joined #openstack-ironic03:14
*** sdake_ has quit IRC03:15
*** ChrisAusten has joined #openstack-ironic03:15
*** sdake has joined #openstack-ironic03:17
*** jaybeale has quit IRC03:20
*** jaybeale has joined #openstack-ironic03:21
*** MattMan has quit IRC03:43
*** MattMan has joined #openstack-ironic03:43
*** jaybeale has quit IRC03:46
*** vmud213 has joined #openstack-ironic03:47
*** jaybeale has joined #openstack-ironic03:49
*** yuanying has joined #openstack-ironic03:49
*** amotoki has quit IRC03:58
openstackgerritDavid Edery proposed openstack/ironic: Fix non-deployed nodes due to short nodes.instance_info  https://review.openstack.org/33496704:02
*** Goneri has quit IRC04:03
*** amotoki has joined #openstack-ironic04:06
*** jaybeale has quit IRC04:14
tiendclucas|afk: vdrok: rloo: hi, I've updated some minor, please help to review: https://review.openstack.org/#/c/328168/04:15
tiendclucas|afk: vdrok: rloo: thanks in advance04:16
*** chlong has joined #openstack-ironic04:20
*** amotoki has quit IRC04:23
*** links has joined #openstack-ironic04:24
*** ametts has quit IRC04:26
*** ametts has joined #openstack-ironic04:26
*** sdake has quit IRC04:27
*** sdake has joined #openstack-ironic04:27
*** amotoki has joined #openstack-ironic04:29
*** amotoki has quit IRC04:38
*** garthb_ has joined #openstack-ironic04:40
*** sukhdev has joined #openstack-ironic04:41
*** amotoki has joined #openstack-ironic04:49
*** ametts has quit IRC05:01
*** amotoki_ has joined #openstack-ironic05:29
*** vmud213 has quit IRC05:29
*** vmud213 has joined #openstack-ironic05:29
*** amotoki has quit IRC05:32
*** ChubYann has quit IRC05:35
*** ametts has joined #openstack-ironic05:37
openstackgerritYuiko Takada Mori proposed openstack/ironic: Follow-up patch of cbe51c6a10d1fa68c7082af67827d8c0d8967fe3  https://review.openstack.org/33865805:37
*** sdake_ has joined #openstack-ironic05:39
*** sdake has quit IRC05:42
*** fragatina has joined #openstack-ironic06:01
*** fragatin_ has joined #openstack-ironic06:02
*** fragatina has quit IRC06:05
*** rcernin has joined #openstack-ironic06:09
*** hshiina has joined #openstack-ironic06:11
*** garthb_ has quit IRC06:12
*** mjura has joined #openstack-ironic06:16
*** aNupoisc has joined #openstack-ironic06:22
*** mjura has quit IRC06:23
*** mjura has joined #openstack-ironic06:24
*** phuongnh has quit IRC06:28
*** phuongnh has joined #openstack-ironic06:29
*** makowals has joined #openstack-ironic06:38
*** Romanenko_K has joined #openstack-ironic06:40
*** aNupoisc has quit IRC06:40
*** chlong has quit IRC06:42
*** pcaruana has joined #openstack-ironic06:51
*** baoli has quit IRC06:51
*** hoangcx has quit IRC06:54
*** hoangcx has joined #openstack-ironic06:54
*** tesseract- has joined #openstack-ironic06:59
*** ifarkas has joined #openstack-ironic07:09
*** phuongnh has quit IRC07:33
*** sukhdev has quit IRC07:33
*** mrda is now known as mrda_afk07:37
*** makowals has quit IRC07:39
*** makowals has joined #openstack-ironic07:39
yibovdrok, no problem07:41
*** lucas|afk is now known as lucasagomes07:43
lucasagomesmorning all07:43
*** jpich has joined #openstack-ironic07:44
lucasagomestiendc, thanks07:48
aarefievmorning all!07:50
aarefievmorning lucasagomes07:50
Romanenko_Khi!07:50
ifarkasmorning all!07:50
tiendcmorning, lucas07:50
tiendcmorning all07:50
aarefievmorning Romanenko_K ifarkas tiendc07:50
*** athomas has joined #openstack-ironic07:51
openstackgerritAnton Arefiev proposed openstack/ironic-inspector: WIP: Tempest: Use tempest call_until_true  https://review.openstack.org/33095007:51
*** zzzeek has quit IRC08:00
*** zzzeek has joined #openstack-ironic08:00
*** rcernin has quit IRC08:02
*** amoralej has joined #openstack-ironic08:05
*** yuanying is now known as yuanying_off08:06
*** sdake_ has quit IRC08:07
amoralejHi, we are having an issue with ironick-inspector unit test ironic_inspector.test.unit.test_main.TestInit caused by changed in stevedore last commit08:07
amoralejhttps://github.com/openstack/stevedore/commit/f68325fc481a0cc2a9aa9e8b1701cdb092463b4908:07
amoralejhttps://paste.fedoraproject.org/388848/14679652/08:08
*** sdake has joined #openstack-ironic08:08
*** sdake has joined #openstack-ironic08:08
amoralejin the past NamedExtensionManager raised KeyError when called with inexisting names, now it's not doing that08:09
openstackgerritDavanum Srinivas (dims) proposed openstack/ironic: [WIP] Testing latest u-c  https://review.openstack.org/31844008:10
*** loki_ has joined #openstack-ironic08:11
*** jralph has quit IRC08:15
*** hoangcx has quit IRC08:15
*** MattMan has quit IRC08:15
*** appu103_ has joined #openstack-ironic08:16
*** appu103_ is now known as aparnav08:16
*** hoangcx has joined #openstack-ironic08:17
*** jralph has joined #openstack-ironic08:18
*** MattMan has joined #openstack-ironic08:19
openstackgerritMerged openstack/ironic-inspector: Make Ironic variables visible inside exercise.sh  https://review.openstack.org/33872708:22
*** rcernin has joined #openstack-ironic08:22
*** daemontool has joined #openstack-ironic08:23
*** ccamacho is now known as ccamacho|afk08:27
*** derekh has joined #openstack-ironic08:27
*** hshiina has quit IRC08:35
*** edand has joined #openstack-ironic08:38
amoralejdtantsur, could i get some help with https://bugs.launchpad.net/ironic-inspector/+bug/1600141 ?08:38
openstackLaunchpad bug 1600141 in Ironic Inspector "test_init_failed_processing_hook fails after https://review.openstack.org/#/c/337043/" [Undecided,New]08:38
*** electrofelix has joined #openstack-ironic08:43
*** stendulker has joined #openstack-ironic08:45
dtantsurMorning Ironic09:01
dtantsuramoralej, hi! ouch, that's nasty :(09:03
dtantsuramoralej, I think we should switch to whatever callback they now provide09:04
openstackgerritAnton Arefiev proposed openstack/python-ironicclient: Fix py35 unit tests  https://review.openstack.org/33877709:04
dtantsuramoralej, seems like we need to provide on_missing_entrypoints_callback to __init__ instead and process failed extensions there09:05
amoralejyes dtantsur, that looks like a good idea09:05
*** sambetts|afk is now known as sambetts09:05
sambettsMorning all09:05
dtantsurmorning sambetts, aarefiev, ifarkas, lucasagomes, tiendc and everyone else09:05
sambettso/ dtantsur09:05
dtantsuramoralej, wanna put up a patch? I'll happily review it09:05
aarefievmorning dtantsur sambetts09:06
amoralejok, i'll need some help dtantsur09:06
dtantsursure09:06
amoralejbut what'd be the expected return from the callback function?09:07
dtantsuramoralej, to match the current behaviour, we should use sys.exit there. we can think of a better approach later, lets just fix the issue09:07
amoralejok09:08
dtantsurbtw if you don't have cycles to work on it, I can handle this issue, no problem with it09:08
amoralejno, i have some time09:09
dtantsurawesome, thanks!09:09
*** aswadr_ has joined #openstack-ironic09:17
*** hkominos has quit IRC09:27
*** hkominos has joined #openstack-ironic09:27
hkominosGood morning ironic09:27
*** ccamacho|afk is now known as ccamacho09:27
sambettso/ hkominos09:28
openstackgerrityuyafei proposed openstack/ironic: Correct reraising of exception  https://review.openstack.org/33943109:29
*** sdake has quit IRC09:30
*** athomas has quit IRC09:31
dtantsursambetts, ifarkas, aarefiev, folks, I wonder if we can/should land https://review.openstack.org/#/c/316801/ without the ironic spec landing...09:33
dtantsurwe're slowly approaching various feature freezes09:33
sambettsdtantsur: merging the inspector side won't affect the current Ironic functionality so I think that's ok09:34
dtantsuryeah, my only concern is what if ironic folks don't agree with the direction. won't we end up with a useless API feature?09:34
dtantsureven though I don't see other ways of implementing it09:34
sambettsdtantsur: tbh I personally think the inspector side of the change is pretty solid, the only part of the Ironic side that I think is controversial is how its configured right?09:36
dtantsuryeah09:36
dtantsurI'll remove my -2 and let the reviewers decide on it09:36
sambettsawesome :D09:37
amoralejdtantsur, it will work if the callback returns KeyError also09:38
*** athomas has joined #openstack-ironic09:38
amoralejand i see this cleaner09:38
amoralejinstead of doing the sys.exit, what do you think?09:38
dtantsuramoralej, to make it even cleaner lets create our own exception class derived from KeyError. then catch it and use sys.exit in the main() function. wdyt?09:39
amoralejok09:40
*** mgould|afk is now known as mgould09:45
mgouldmorning Ironic!09:45
amoralej:wq!09:47
dtantsuramoralej, you cannot exit vim, sorry09:50
amoralej:DD09:50
amoralejdual screen...09:51
dtantsurtrue, I also write at least one vim command per week in irc for the same reason :)09:51
aarefievdtantsur: big +1, it's for sure speed up all work, and it's very desired feature09:55
dtantsurcool09:56
aarefievdtantsur: btw, was confused, would be possible use deploy ramdisk ids by default, w/o duplicating inspection one?09:57
aarefievin driver info09:57
dtantsurI'm not against it, but potentially that could be different ramdisks09:58
dtantsure.g. a lightweight deploy ramdisk and a heavy inspection one09:58
aarefievdtantsur: sure, just won't copy/paste fields in case same image10:00
*** hoangcx has quit IRC10:01
dtantsuraarefiev, please leave the comments, lets continue on the spec10:01
* dtantsur does not remember his spec already :D10:02
aarefievdtantsur: actually I did iirc10:02
dtantsurawesome, I'll check it10:02
*** vsaienk0 has quit IRC10:06
*** vsaienk0 has joined #openstack-ironic10:09
ifarkasdtantsur, the manage_boot change looks good to me, although I was thinking why do we want to land it if the changes are missing from ironic10:12
ifarkasare we expecting use cases outside ironic?10:13
amoralejdtantsur, i have the patch ready but i just realized that the commit that introduced  on_missing_entrypoints_callback is not tagged yet10:14
amoralejso it will not pass the gate10:14
amoralejhttps://github.com/openstack/stevedore/commit/6f81f6f1c1b2e9efaa7e90a3f8a6b57007a8d48610:14
amoralejright?10:14
*** rcernin has quit IRC10:15
dtantsurifarkas, maybe?10:21
dtantsuramoralej, ooops. yes, you're right. lets post the patch and wait for the release10:22
amoralejok10:22
dtantsurit's an interesting issue, actually. #rdo can potentially get even more serious problem due to that10:23
dtantsuramoralej, but wait, if we derive our exception from KeyError, it should be backward compatible, no?10:24
dtantsurI think we should be fine10:24
amoralejno, problem is that calling NamedExtensionManager with on_missing_entrypoints_callback arg fails10:25
dtantsurwut, did they just introduce it?10:25
amoralejyes10:25
amoralejabout two weeks ago10:25
dtantsurdamn.. moving to #openstack-oslo10:26
*** yonglihe has quit IRC10:33
*** yonglihe has joined #openstack-ironic10:35
openstackgerritAlfredo Moralejo proposed openstack/ironic-inspector: Add callback function to manage missing hooks  https://review.openstack.org/33945710:39
amoralejdtantsur ^ i guess it'll fail in gates but at least we can review it10:41
dtantsurthanks10:41
*** amotoki_ has quit IRC10:42
*** hoangcx has joined #openstack-ironic10:56
*** loki_ has quit IRC11:00
openstackgerritImre Farkas proposed openstack/ironic: DRAC RAID configuration  https://review.openstack.org/17631911:00
*** stendulker has quit IRC11:05
*** rcernin has joined #openstack-ironic11:08
openstackgerritImre Farkas proposed openstack/ironic: Delete bios_wsman_mock.py from DRAC driver  https://review.openstack.org/33947111:10
openstackgerritDmitry Tantsur proposed openstack/python-ironic-inspector-client: Support passing manage_boot argument in Python API  https://review.openstack.org/33947911:27
mat128|afksambetts: about https://review.openstack.org/#/c/327891/8/imagebuild/coreos/coreos-oem-inject.py11:28
*** mat128|afk is now known as mat12811:28
mat128JayF uses the COREOS_VERSION later for the .imagecache11:29
openstackgerritMerged openstack/ironic: Delete bios_wsman_mock.py from DRAC driver  https://review.openstack.org/33947111:29
mat128sambetts: so we need to fetch it anyway (unless we want to change that part too)11:29
mat128sambetts: and I guess having current both for "version fetching" and for "artifacts fetching" could lead to a race condition issue during a CoreOS release (unlikely, but yeah..)11:30
openstackgerritMathieu Mitchell proposed openstack/ironic-python-agent: Use latest CoreOS stable when building  https://review.openstack.org/32789111:32
vsaienk0morning dtantsur, sambetts, aarefiev, mat128 and all Ironic'ers!11:33
mgouldmorning vsaienk011:33
dtantsurmorning vsaienk0, mat12811:34
mat128morning vsaienk0, mgould, dtantsur11:34
mat128friday!11:34
dtantsur\o/11:34
mat128sambetts, lucasagomes and dtantsur: let me know what you think of https://review.openstack.org/#/c/327891/11:34
mat128I hope to have this merge before my 3 week PTO ;P11:35
sambettshmm, interesting :/ WRT the image cache I'm not sure if we need to cache each different version, as we now have no way to override that version to go back to older ones, so just caching a single version of each file we need should be fine, and only overriding them if the version of the server is different to that we already have11:35
dtantsurI'm slightly worried about always using the latest version, to be honest..11:35
dtantsurI can expect them breaking us from time to time11:35
sambettsdtantsur: latest stable11:35
dtantsurwhatever11:36
dtantsurwe don't run our gates against their git ;)11:36
mat128I remember this discussion when JayF was around and his point was I'd rather fail early11:36
mgouldwhich images are we caching?11:36
mat128mgould: the coreos binaries11:37
mgouldcool, thanks11:37
dtantsurmat128, also, why not use https://stable.release.core-os.net/amd64-usr/current/ instead of detecting versions?11:37
sambettsdtantsur: thats the dicussion we're just having ;)11:37
dtantsurah!11:37
lucasagomesmat128, will do in a moment11:37
mat128=)11:37
dtantsurwhat are the cons of doing it?11:37
mat128dtantsur: we "need" the real version number to build the image cache11:38
mat128sure on an ephemeral slave it does matter because everything is gone after it ran11:38
mgouldcoreos releases a new version and all our stuff breaks, I'm guessing11:38
mat128the alternative is to pin to a version and have a proposal bot suggest new versions as they come out11:38
mgould+1 to pin + bot11:38
* mat128 thinks we're not going to complete this by today lol, sorry Jay11:39
sambetts^ that sounds like a better way to track it, because it will also mean that we have to do a merge each time there is a version change which will trigger the post job11:39
mat128but yeah, that would be a way of protecting ourselves from COreOS releases while tracking them11:39
dtantsurI'd also slightly prefer a bot (if it's possible). this is in line with how openstack handles external dependencies11:39
mat128sambetts: yes11:39
mat128would you guys agree to us "pinning" it to the current version so we at least leverage the work we did?11:40
dtantsurdefinitely11:40
sambettsif it works I have no problem with it11:40
mat1281010.6.011:40
mat128ok11:40
mat128then we can work the proposal bot thing later11:40
mat128I know how to do it (did it downstream for maven dependencies) but it's project-config stuff11:40
dtantsuryep. for now we can do it manually from time to time.11:40
dtantsursambetts, I've heard the was a new release of tinycorelinux too ;)11:41
mgouldmat128: how does the proposal-bot thing work?11:42
mat128jenkins job that runs in a "periodic" zuul pipeline11:42
mat128that job clones a project, modifies files and literally does a "git review"11:42
sambetts7.1 to 7.2 and we're pulling from 7.x so we should get it automatically11:43
mat128the periodic pipeline is just like a cron11:43
mat128sambetts: oh sh11:43
mgouldoh, cool11:43
mat128sambetts: same issue then?11:43
mat128the other thing is that by pulling the latest (whether 7.x or "current" CoreOS), we produce no build unless we change code11:43
*** hoangcx has quit IRC11:43
mat128this happens frequently enough here though11:44
mgouldmat128: add "echo $coreos_version >> .coreos_version" to the build script, so a file has changed?11:44
*** ccamacho is now known as ccamacho|lunch11:44
mat128mgould: I mean, you would have to submit that file through a gerrit review and have it merge11:45
mat128to re-run the "post" pipeline11:45
sambettsmat128: I'd hope by only tracking auto pulling minor versions we should be ok with tinycore eg. 7.1 -> 7.2 should technically break anything11:45
mgouldoh, sorry, I thought we were talking about the hypothetical bot11:46
sambettscurrent in coreos might go major to major though11:46
mgouldsambetts: does tinycore use semver?11:46
mat128sambetts: shouldn't, yeah11:46
sambettsshouldn't *11:46
mgouldbut still might, because software is difficult11:46
mat128mgould: in our current case, the bot would submit a change to the build script (but yeah an external file would be much more practical)11:46
*** tiendc has quit IRC11:46
mat128mgould: in fact, the bot could download and commit this file: http://stable.release.core-os.net/amd64-usr/current/version.txt11:47
mat128and we would read from it11:47
mgouldmat128: I like it11:47
sambettstbh along side all of this we have no control on the versions of the packages either the coreos or tinycore builder pulls right, because thats all down to the package managers11:47
mgouldso then the bot script is just "wget $url; git diff | git review"11:48
sambettsI like that ^11:48
mgouldsambetts: bah11:49
mgouldso we can still be using a given version of coreos/tinycore, and still have our images change out from under us?11:49
sambettsmgould: nice and straight forward: git clone ironic-python-agent && cd ipa/imagebuild/coreos && wget version.txt && git commit -m "New CoreOS Version" && git review11:50
sambettsmgould: potentially I guess yes, and I think we've seen this before with a version when a version of a package changed11:50
mat128sambetts: pretty much that, yes :)11:50
mat128mgould: yes images can change under our feet11:51
mat128just like Ubuntu publishes new packages11:51
mat128we trust the maintainers to not break our systems11:51
*** lucasagomes is now known as lucas-hungr11:51
*** lucas-hungr is now known as lucas-hungry11:51
mat128on that side11:51
openstackgerritMathieu Mitchell proposed openstack/ironic-python-agent: Bump CoreOS to 1010.6.0 (last current stable)  https://review.openstack.org/32789111:52
mat128^ pinned to 1010.6.011:52
openstackgerritAlfredo Moralejo proposed openstack/ironic-inspector: Add callback function to manage missing hooks  https://review.openstack.org/33945711:53
mgouldsambetts: oh, ofc, git-commit will fail if nothing has changed11:54
openstackgerritVasyl Saienko proposed openstack/ironic: WIP: Poll Neutron port status.  https://review.openstack.org/33948911:55
dtantsuramoralej, do you plan on proposing one more patch (skipping the test which failed)?11:56
amoralejyes11:56
dtantsurthanks11:56
amoraleji'm on it...11:56
openstackgerritAlfredo Moralejo proposed openstack/ironic-inspector: Remove test_init_failed_processing_hook test  https://review.openstack.org/33949112:06
*** aparnav has quit IRC12:07
amoralejdtantsur, ^ should pass with old and new releases12:07
*** alaski is now known as lascii12:07
dtantsuralready responded :)12:07
*** appu103_ has joined #openstack-ironic12:08
amoralejdidn't know that skipTest, good to learn12:08
amoralejthanks12:08
openstackgerritMathieu Mitchell proposed openstack/ironic-python-agent: Bump CoreOS to 1010.6.0 (last current stable)  https://review.openstack.org/32789112:11
mat128mgould, sambetts: did half the work for the proposal bot stuff, we have an entrypoint to bump the pinning and we already read from the file :)12:11
mat128^12:12
*** amotoki has joined #openstack-ironic12:12
mgouldmat128: \o/12:12
*** dprince has joined #openstack-ironic12:16
sambettsmat128: I wonder if we should allow COREOS_VERSION to be overridden through env vars e.g. COREOS_VERSION=${$COREOS_VERSION:-get_coreos_version(COREOS_VERSION_FILE)} ??12:20
mat128sambetts: sure why not :)12:21
mat128thats python though12:21
sambettsmat128: oh yeah :/ hmmm12:21
mat128(just as easy)12:21
mat128ill do it12:21
sambettsawesome, just thinking that then people can pin their own version if they want12:22
*** joprovost has joined #openstack-ironic12:22
openstackgerritMathieu Mitchell proposed openstack/ironic-python-agent: Bump CoreOS to 1010.6.0 (last current stable)  https://review.openstack.org/32789112:23
mat128sambetts: done12:23
mat128yes, good idea :)12:23
mat128the only thing is if you want a very old version12:23
mat128the URLs might have changed12:23
mat128and/or signature12:23
mat128but if thats what you waht you might as well modify the .py12:24
mat128you're not going to pass a signature via env var ;)12:24
sambettstrue true :P12:24
*** trown|outtypewww is now known as trown12:25
openstackgerritAlfredo Moralejo proposed openstack/ironic-inspector: Skip test_init_failed_processing_hook test  https://review.openstack.org/33949112:25
*** joprovost has quit IRC12:27
*** joprovost has joined #openstack-ironic12:28
*** jcoufal has joined #openstack-ironic12:32
openstackgerritMathieu Mitchell proposed openstack/ironic-python-agent: Bump CoreOS to 1010.6.0 (last current stable)  https://review.openstack.org/32789112:33
mat128sambetts, mgould: another one :)12:33
mat128sorry Miles I had not seen your comments for my previous one, addressed them now12:33
*** appu103_ has quit IRC12:35
sambettsmat128: heh I was just moving mgould's comment over to the new patchset, but I was just going to suggest change all the logic to "if ("%s="%key) in line: return line.strip().split("=")[1]12:35
openstackgerritOpenStack Proposal Bot proposed openstack/ironic-python-agent: Updated from global requirements  https://review.openstack.org/32854012:36
*** rbudden has joined #openstack-ironic12:37
dtantsursambetts, up for a quick review of https://review.openstack.org/#/c/339491/ ? this has potential of breaking our gate as soon as new stevedore version is released12:39
sambettsdtantsur: cool, what a pain12:42
sambettsamoralej: thanks for the patch12:42
amoralejglad to help sambetts12:43
*** mjura has quit IRC12:43
openstackgerritMerged openstack/ironic-inspector: Skip test_init_failed_processing_hook test  https://review.openstack.org/33949112:48
* mgould googles stevedore12:51
mgouldman, *another* openstack project I didn't know about12:51
mgouldI learned about three more on Wednesday morning12:52
dtantsurlol12:52
dtantsurthere are dozens of them12:52
*** amoralej is now known as amoralej|lunch12:53
*** lucas-hungry is now known as lucasagomes12:53
mgoulddtantsur: mostly with names that only make sense once you know what they do12:53
mat128haha12:53
dtantsurtrue12:53
mat128stevedore never makes sense12:53
mgouldyeah, I'm still mystified12:53
mat128ehh12:53
mat128never mind12:53
mgouldoh, someone who loads and unloads things12:54
mat128"Manages dynamic plugins" (readme)12:54
mat128and "a person employed, or a contractor engaged, at a dock to load and unload cargo from ships."12:54
mgouldand who has been made obsolete by containerization? :-)12:54
mat128Haha12:54
*** vmud213 has quit IRC12:57
*** vmud213 has joined #openstack-ironic12:58
*** joprovost has quit IRC12:58
mgouldsambetts: nice :-)12:59
jrollhappy friday everyone \o/13:05
mat128sambetts, mgould, dtantsur : that might be of interest to you https://review.openstack.org/#/c/339543/13:05
mat128happy friday jroll :D13:05
jrollmgould: stevedore was taken over by openstack, used to be external13:06
jrollmgould: also, favorite openstack repo I've found lately https://github.com/openstack/ailuropoda/13:06
mgouldwat13:06
dtantsurwow13:06
dtantsurwe need to support it in tripleo13:07
mgouldjroll: is that like GNU Hello World? A project that does something stupid in a way that complies with all the project standards?13:07
mgouldsad pandas as a service13:07
jrollmgould: I have no idea, this is all I know13:07
jrollbut I love it13:08
mgouldwow, that is a very sad panda13:08
mgouldhttps://github.com/openstack/ailuropoda/blob/master/images/v1/sad_panda01.png13:08
jrollit has to be a hello world thing13:09
jrollthey have an irc channel even13:09
mgoulda friend of mine wrote a kitten-pics-as-a-service tool for motivating writers: http://writtenkitten.co/13:09
mat128looks like a hello world / skeleton project13:09
mat128mgould: ^ thats actually motivating13:10
mat128I wonder if I can write specs in that editor :)13:10
mgouldmat128: definitely worth a try13:10
jrollI still can't decide if this is serious13:11
mgouldthere's also https://placekitten.com/, which is useful for web developers13:11
mgouldjroll: writtenkitten? I'd say it's about 50% serious13:11
mgouldlike, it started as a joke, but people use it for real13:11
mat128:P13:12
jrollmgould: the sad panda thing13:12
jrollit can't be13:12
jrollbut it's very convincing13:12
jrolle.g. this commit message https://github.com/openstack-infra/project-config/commit/c9fdb13:13
mat128jroll: we internally train new developers by having them create projects similar to this, with full CI builds and (almost) staging/production deployment13:13
mat128leaves a bunch of projects / jenkins jobs behind if they dont clean it up, but at least they get to know the internals13:13
jrolloh, neat13:13
jrollthis is chef people13:13
* jroll suspects troll13:13
mat128jroll: oh god13:14
mgouldmat128: that's an excellent idea13:14
mat128the commit message13:14
mgouldwow13:14
mat128doing it is the best way to understand it13:14
mgouldyep13:14
jrolloh no, it's actually openstack chef people13:14
jrollhuh13:14
mat128ah13:14
mgouldmat128: are there docs for that publicly available?13:15
mgouldsounds like a process I should go through myself13:15
mat128no docs :)13:15
mat128we want them to learn, not follow a step by step howto13:15
mgouldwork it out yourself? heh13:15
mat128they have to go ask questions13:15
mgouldright, gotcha13:15
*** links has quit IRC13:15
mat128but our latest CI stack is a copy of what openstack has13:15
mat128gerrit/zuul/jjb and friends13:15
mgouldoh, right13:15
mgouldinternal CI, of course13:16
mat128yup13:16
dtantsuromg, this repo seem to have 5 cores according to https://wiki.openstack.org/wiki/Ailuropoda13:16
*** piet has joined #openstack-ironic13:16
mat128mgould: thats a perfect segue to the proposal bot changes :)13:16
mat128https://review.openstack.org/#/c/339543/13:16
mgouldyep, looking now13:17
mat128Ailuropoda reminds me of what an ops of our did for HTTP errors13:17
mat128he returned ascii kittens13:17
mgoulddtantsur: sounds like an excellent way of getting into the s3kr1t core party at summit :-)13:17
dtantsur\o/13:18
mgouldmaybe I'll recolour pixieboots and submit him as an alternate sad-panda image13:18
jrollmgould: it's the ex- openstack-chef PTL, I suspect he can get in there already :P13:18
TheJuliaGood morning13:19
mat128o/ TheJulia13:19
mgouldjroll: Matthew 25:2913:19
*** baoli has joined #openstack-ironic13:19
mgouldmorning TheJulia13:20
mat128mgould: nice quote13:20
jrollmgould: oh my, the parallels with the current state of the world :(13:20
*** baoli_ has joined #openstack-ironic13:20
openstackgerritKyrylo Romanenko proposed openstack/python-ironicclient: Add tests for provision state commands  https://review.openstack.org/33729613:20
mgouldjroll: yep :-(13:21
* jroll reads context13:21
jrollah, this story13:22
jrollhuh.13:22
* jroll ponders13:23
mgouldyeah, I didn't consider the context13:23
mgouldor even remember it, TBH13:23
dtantsurjroll, please approve https://review.openstack.org/#/c/328540/ (or should I just approve it the next time?)13:24
jrolldtantsur: done, I don't mind single +2 for those13:24
dtantsurk13:24
jrolldtantsur: but I'm also just bad at reviewing stable things :/13:25
*** baoli has quit IRC13:25
dtantsurthey easily fall out of sight13:25
jrolldtantsur: trying to figure out v2 api meeting, is 1800 utc too late for you I guess?13:27
openstackgerritMathieu Mitchell proposed openstack/ironic-python-agent: Bump CoreOS to 1010.6.0 (last current stable)  https://review.openstack.org/32789113:27
mat128mgould: done :) ^13:27
jrolldtantsur: oh, found a 1600 slot if that's better13:27
mat128hopefully the last changeset, I have to drive to the office now13:27
jrolltrying to find one that's good for you and for pacific time folks13:28
* mat128 afks for ~30m13:28
*** mat128 is now known as mat128|afk13:28
jrollugh, but that conflicts for me13:28
jrollT_T13:28
jrollthis is hard13:28
mgouldmat128|afk: +1 from me13:28
dtantsurjroll, 18:00 is doable, but I'd definitely prefer anything earlier13:29
jrolldtantsur: okay, I found a 1500 that I think should be okay for everyone13:30
jrollthat's 8am pacific13:30
dtantsur1500 what day?13:30
*** cdearborn has joined #openstack-ironic13:31
jrolldtantsur: tuesday13:31
dtantsurmeaning 1700 my time. seems fine13:31
jrollcool, thanks13:32
*** ccamacho|lunch is now known as ccamacho13:34
*** sabeen has joined #openstack-ironic13:34
*** dansmith is now known as superdan13:39
*** wajdi has quit IRC13:40
*** PollyZ has joined #openstack-ironic13:47
lucasagomesjroll, morning, are you sending it to the ML ?13:47
jrolllucasagomes: I will be13:47
lucasagomescool :-)13:47
jrollturns out the generated ical file is out of date, so that time conflicts13:47
jroll:(13:47
*** cdearborn has quit IRC13:48
lucasagomes:-/ maybe on wednesday>13:48
lucasagomes?*13:49
dtantsurlucasagomes, jroll, have you seen https://review.openstack.org/#/c/339550/ ? :(13:49
lucasagomesdtantsur, urgh, nop :-/13:49
sambetts-,-13:50
jrolllol13:50
mgouldlucasagomes, jroll, dtantsur: doodle poll?13:50
jrollmgould: I mean...13:50
jrollmost of it is having an open meeting room13:50
mgouldOIC13:50
mgoulddtantsur: aaaaargh13:51
mgouldnobody's replied to my ML post about that, AFAICT13:51
*** joprovost has joined #openstack-ironic13:52
jrollmgould: http://i.imgur.com/xXv0PGf.png13:52
jrolland also trying to keep it at a time that works for west coast US and europeans13:52
jrollpretty limiting13:52
mgouldzomg, that's a lot of meetings13:52
* sambetts is looking forward to July 19th, I should be able to host webex meetings that work on Linux then 13:52
mgouldyeah, EU/NA is hard13:52
* jroll will still refuse to webex as much as possible :P13:52
sambettshehe13:53
mgoulddtantsur, lucasagomes: one person in #openstack-nova said "if you think extra_spec_ops is bad, you should see $other_module"13:54
mgouldI have mercifully forgotten which one $other_module was13:54
dtantsurlol13:54
lucasagomes:D13:54
jrollmgould: it was the compute capabilities filter that uses it :P13:55
*** Goneri has joined #openstack-ironic13:55
*** amoralej|lunch is now known as amoralej13:59
*** vmud213 has quit IRC14:01
mgouldjroll: ah, thanks14:02
* mgould may take a look when he's feeling stronger14:02
*** joprovost has quit IRC14:02
*** joprovost has joined #openstack-ironic14:03
* jroll wonders how terrible monday 1500 would be14:06
jrollironic api, ironic/neutron, ironic meetings all in a row :(14:06
openstackgerritKyrylo Romanenko proposed openstack/python-ironicclient: Add tests for provision state commands  https://review.openstack.org/33729614:06
*** hkominos has quit IRC14:06
jrollI think it'll need to be 1800 tuesday dtantsur :/14:07
dtantsurlucasagomes, it seems like we'll need operators for extra specs in ironic-lib after all14:07
dtantsur:(14:07
dtantsurjroll, I'll try to be there, but I guess it won't be an end of the world if I have to skip some, right?14:07
dtantsurI'm not the most active person in the V2 process14:07
jrollyeah, indeed14:07
jrollright14:07
lucasagomesdtantsur, yeah :-/14:08
mgoulddtantsur, lucasagomes: ugh, this was not the effect I was hoping for14:08
openstackgerritCaio Oliveira proposed openstack/python-oneviewclient: [WIP] Creating certificate manager for OneView  https://review.openstack.org/33499214:08
jrolldtantsur: keep in mind with the extra specs stuff that harlowja is the PTL, and he was okay with it14:08
* jroll finds popcorn14:08
lucasagomesjroll, what do you think? Should we have it in ironic-lib or wait for oslo?14:08
lucasagomesor maybe just get the ones in inspector and put in ironic-lib14:09
jrolllucasagomes: we shouldn't abandon hope for oslo just because one core dislikes it14:09
lucasagomesand use that on both projects ironic and inspector?14:09
jrollso let's wait and see14:09
dtantsurwe can wait until library freeze, then figure out14:09
* dtantsur gets popcorn as well14:09
lucasagomesjroll, sure, in any case the oslo one will be backward compatible with nova... that's why I was thinking in having it in ironic-lib to unblock us14:09
lucasagomesonce it's merge in oslo we migrate14:10
*** mtanino has joined #openstack-ironic14:10
jrolllucasagomes: sure, but let's give it a day or two14:10
jrollthe oslo meeting is monday morning14:10
lucasagomesjroll, cool14:10
jrollsurely this will get attention there and we can decide then :()14:10
jroll:)14:10
lucasagomesfair enough14:10
*** mbound has joined #openstack-ironic14:11
*** derekh is now known as derekh_gone_to_g14:13
*** PollyZ has quit IRC14:14
openstackgerritMerged openstack/ironic-python-agent: Updated from global requirements  https://review.openstack.org/32854014:15
*** PollyZ has joined #openstack-ironic14:17
*** hoangcx has joined #openstack-ironic14:19
mgouldjroll: https://review.openstack.org/#/c/339576/14:20
*** hoangcx has quit IRC14:20
jrollmgould: so my concern with changing this is wondering how many deployments it will break14:21
*** PollyZ has quit IRC14:22
mgouldjroll: AFAICT we haven't changed any behaviour yet14:23
jrollmgould: yeah, not that patch specifically, just saying14:23
mgouldmy patch ^^^ alters the tests to (hopefully) test the intention better14:24
mgouldharlowja's "use a parser generator" patch also shouldn't change behavior14:24
jrollthe assumed intention :)14:24
mgouldyeah, true14:24
mgouldit's definitely unsafe to replace the implementation wholesale if we don't understand what it's trying to do14:25
openstackgerritMathieu Mitchell proposed openstack/ironic-python-agent: Bump CoreOS to 1010.6.0 (last current stable)  https://review.openstack.org/32789114:25
*** sabeen1 has joined #openstack-ironic14:25
mat128|afkmgould, sambetts: ^ pep8 fixed... :(14:26
*** sabeen has quit IRC14:27
mgouldpesky pep814:28
mgould+1ed :-)14:29
mgouldbrb packing14:29
*** PollyZ has joined #openstack-ironic14:30
openstackgerritLucas Alvares Gomes proposed openstack/ironic: Remove IBootOperationError exception  https://review.openstack.org/33912714:32
openstackgerritYossi Ovadia proposed openstack/ironic-python-agent: Handle diskless hardware connected to remote iscsi  https://review.openstack.org/32780714:34
*** joprovost1 has joined #openstack-ironic14:34
*** mat128|afk is now known as mat12814:35
*** PollyZ has quit IRC14:35
openstackgerritAndre Aranha proposed openstack/python-oneviewclient: WIP Extend to Server Profile Connections  https://review.openstack.org/33411914:35
*** joprovost has quit IRC14:36
*** joprovost1 is now known as joprovost14:36
dtantsurThe requested URL /calendars/ironic-v2-api-meeting.ics was not found on this server.14:38
dtantsurI guess eavesdrop is not updated yet14:38
*** wajdi has joined #openstack-ironic14:38
dtantsurjroll, fyi: I'll definitely skip the 1st V2 meeting for personal reasons (aka wedding anniversary)14:40
*** PollyZ has joined #openstack-ironic14:40
jrolldtantsur: sounds like a good reason \o/14:40
dtantsuryep :)14:40
jrolland yeah, I need to bribe people to merge the meetings patch: https://review.openstack.org/#/c/339211/14:40
thiagopgood morning folks14:41
mgouldmorning thiagop14:42
thiagopdtantsur: are you married? really?14:42
dtantsurweeelll.... yes :)14:42
thiagopdtantsur: that corroborate my theory that you're older than appear14:42
mgouldthiagop: and for at least one year :-)14:43
*** edand has quit IRC14:43
dtantsurnot substantially probably :)14:43
thiagopmgould: lol14:43
*** piet has quit IRC14:43
dtantsurlol, indeed14:43
mgouldany similarities between dtantsur and 15th-century portraits of Prince Dmitry the Preternaturally Productive are purely superficial14:44
openstackgerritLucas Alvares Gomes proposed openstack/ironic: Mock global variables  https://review.openstack.org/33866514:44
dtantsurLOOOOL14:45
* lucasagomes updates a tiny thing14:45
*** PollyZ has quit IRC14:45
thiagopThat I wanna google14:45
*** links has joined #openstack-ironic14:45
mgouldthiagop: alas, I made him up14:46
dtantsurnow I wonder what your reaction is when you know it's 8th anniversary :D14:46
lucasagomesoh wow14:46
mgouldthere was a King Ladislav the Posthumous of Bohemia, though14:46
mgoulddtantsur: gosh14:46
mgouldalso, congratulations!14:46
thiagopholy cow! I need to see more of the world...14:46
lucasagomesdtantsur, indeed, congrats :-)14:47
dtantsur:D14:47
thiagopdtantsur: I'd give you 27 years old, at maximum14:47
*** sabeen1 has quit IRC14:47
dtantsurthiagop, you're right actually, I don't plan to become 28 before this December14:47
thiagopfor the record, devstack-gate should have being setting deploy_callback_timeout to 600 statically? I'm thinking about sending a patch to remove that (or at least make it configurable)14:47
thiagopdtantsur: so you married pretty earlier14:47
dtantsuryep14:48
dtantsur(unfortunately your guess about time-travel is wrong)14:48
*** PollyZ has joined #openstack-ironic14:48
*** rajinir has joined #openstack-ironic14:51
*** piet has joined #openstack-ironic14:53
thiagoprajinir sambetts have you received my e-mail about the 3rd party CI talk?14:53
*** links has quit IRC14:54
sambettsthiagop: Yeah, I'm not sure what to respond with :-P, pretty much everything I want to talk about it covered on there, and I'm more than happy to be an member on the panel14:55
sambettsthiagop: not sure who should MC it though14:56
thiagopsambetts: you can be the moderator, but I'm jusck ack'ing. You can answer when you have time to think about14:56
thiagopjlvillal: around?14:57
*** makowals has quit IRC14:57
sambettsthiagop: I'm totally up for it and you should summit the proposal for it though :)14:57
sambettsbrb14:57
*** sambetts is now known as sambetts|afk14:57
*** PollyZ_ has joined #openstack-ironic14:58
*** PollyZ has quit IRC15:01
*** skramaja has quit IRC15:02
rajinirthiagop: Got it, happy to be part of it. Will reply today15:02
thiagoprajinir: ack ;)15:03
*** derekh_gone_to_g is now known as derekh15:04
*** sabeen has joined #openstack-ironic15:04
*** sdake has joined #openstack-ironic15:10
*** vishwanathj has joined #openstack-ironic15:10
mat128one more core? :) https://review.openstack.org/#/c/327891/15:11
dtantsurI think jroll or JayF should look at ^^^15:11
*** sdake_ has joined #openstack-ironic15:12
*** pcaruana has quit IRC15:13
thiagopsambetts|afk: deployment with Ubuntu user image is taking >10min. I not liking it a lot...15:14
thiagopFolks, if you would be so kind as to review this quick patch that'll allow my CI nodes to be enrolled by devstack, I'd be grateful: https://review.openstack.org/#/c/328451/15:15
jrolldtantsur: mat128: LGTM, waiting for zuul15:15
*** sdake has quit IRC15:15
*** ChrisAusten has quit IRC15:15
thiagopvsaienk0: Thanks for reviewing, btw15:16
lucasagomesmat128, I like the pin thing, good idea15:16
*** sdake_ has quit IRC15:17
*** sdake has joined #openstack-ironic15:18
jrollthiagop: +W15:19
thiagopjroll: thank you, sir15:19
*** jaybeale has joined #openstack-ironic15:20
jrollnp15:21
thiagopThanks dtantsur too15:25
dtantsur:)15:25
*** tesseract- has quit IRC15:27
*** sdake_ has joined #openstack-ironic15:27
sambetts|afkthiagop: I guess that image is 15x the size of the Cirros one15:28
*** sambetts|afk is now known as sambetts15:28
thiagopsambetts: but the problem is some delays on cloud-init and networking process. It shouldn't take that long...15:29
mat128yay!15:29
sambettsthiagop: cloud-init wait 120 seconds for network to come up if thats what you mean? but it does that to guarentee that DHCP has given it an address as I understand it15:30
mat128lucasagomes: thanks :) I already have the proposal bot patch submitted on project-config (https://review.openstack.org/#/c/339543/)15:31
* sambetts wonders if a TinyCore + Glean "user" image for testing would be awesome or not 15:31
TheJuliathiagop: cloud-init also does things like generates new ssh keys15:31
TheJuliasambetts: ++15:31
*** sdake has quit IRC15:32
thiagopI'll try to approach that later, try to optimize...15:32
mat128sambetts: good idea I think15:32
*** ifarkas has quit IRC15:32
thiagopA thing that sambetts might wanna know is that TinyIPA didn't work on our blades15:32
sambettsthiagop: oh interesting, any particualr error?15:33
*** sdake_ has quit IRC15:35
thiagopI think it's network related too, but didn't get a snapshot (and I'm not that familiar with the output)15:35
sambettsthiagop: perhaps missing a nic driver15:35
thiagopI'll try to get a snapshot from the output later today15:35
thiagop(and stop to complain without logs/proof)[15:36
thiagop:)15:36
sambetts;)15:36
sambettsjroll, TheJulia: can we get your input on this, it has 3x+2s but I think we're holding off merging until we get more opinions (/me voted after it had +2s but was too scared to land it ;))15:39
jlvillalthiagop, I am around15:39
* jlvillal needs to fix up his hexchat to highlight his messages better15:40
jrollsambetts: opinions on what?15:40
sambettsjroll, TheJulia: https://review.openstack.org/#/c/338417/15:40
sambettsOOPS :-p15:40
sambettshaha15:40
jrollsambetts: oh, I'm good with the idea, but haven't reviewed the patch15:40
jrollif the code is good, go for it15:40
thiagopjlvillal: please, take a look at this, tell me what you think: https://review.openstack.org/#/c/339604/15:40
sambettsjroll: passing the gate ;)15:41
jrollhehe15:41
*** ChrisAusten has joined #openstack-ironic15:41
TheJuliasambetts: looking now15:41
jlvillalthiagop, will do15:42
sambettsTheJulia: thanks :)15:42
thiagopjlvillal: one line patch15:43
TheJuliasambetts: by chance did you get my rambling that I sent you last night?15:44
sambettsTheJulia: oh yeah, sorry was half way through replying and got distracted XD15:44
TheJuliaheh, no worries15:45
*** sdake has joined #openstack-ironic15:45
*** garthb has joined #openstack-ironic15:50
TheJuliasambetts: +2'ed15:51
sambettsawesome :D15:52
jlvillalthiagop, Looks reasonable to me.15:52
TheJuliaI think it is safe to +a unless jroll really wants to review it.  I think a minor follow-up is needed, bug that is not a big deal15:52
jrollTheJulia: I don't mind it going forward without me, I don't have time to review right now15:54
* TheJulia wonders if this is like a wedding and if anyone has any reasons it should not be merged15:54
mat128jroll: tests passed, investigating that non-voting failure but I've seen it pass with a previous patchset that had only failed pep815:55
mat128https://review.openstack.org/#/c/327891/15:55
jrollmat128: looks like a test timeout to me, probably unrelated15:57
jrollbut I definitely want coreos tests passing15:57
* jroll rechecks15:58
mat128thanks, I'll check the result after lunch15:58
mat128but patch set 14 had that build passing so I'm not worried15:59
jrollah ok16:00
lucasagomesjroll, JayF do you guys bump the "min_command_interval" configuration option downstream ?16:01
lucasagomesthe default value of*16:01
*** jaybeale has quit IRC16:01
jrolllucasagomes: which one is that?16:02
* jroll looks16:02
lucasagomesit's under [ipmi]16:02
lucasagomesthat's the interval between each ipmi command16:02
jrolloh that, no16:02
jrollI don't believe so16:02
jrollyeah, confirmed, it's default16:02
lucasagomescause, downstream there's yet another person telling me they bumped it to ~10s16:02
lucasagomesI've seem a couple of people saying they needed to bump, so I was wondering16:03
mat128lucasagomes: we have seen transient BMC failures but havent got to bump that setting yet16:03
lucasagomesI wonder if we should change that default :-/16:03
jrollyay more bad hardware16:03
lucasagomesmat128, right16:03
lucasagomesjroll, exactly16:03
mat128BMCs should be open, both hardware and software16:03
mat128:D16:03
jrollI mean, I'd be curious to how many people set it higher vs don't16:03
jrollmat128: I wish you the best of luck :)16:04
mat128hahaha16:04
jrollmore seriously, openbmc16:04
lucasagomesmat128, heh totally there are projects around but...16:04
lucasagomesjroll, yeah I don't have that data16:04
lucasagomesit comes sporadically16:04
jrollright16:04
*** ametts has quit IRC16:05
jrollI feel like we just RMA'd any boxes that had really slow IPMI16:05
jrollbut I don't remember offhand16:05
lucasagomesright16:07
*** sabeen has quit IRC16:08
sambettsTheJulia: thanks for +Wing :D16:10
*** Romanenko_K has quit IRC16:12
*** sabeen has joined #openstack-ironic16:13
*** amotoki has quit IRC16:18
*** rcernin has quit IRC16:18
*** pcaruana has joined #openstack-ironic16:21
*** rama_y has joined #openstack-ironic16:21
* mgould -> train -> camping; have a good weekend, everyone!16:23
*** mgould is now known as mgould|afk16:23
openstackgerritLucas Alvares Gomes proposed openstack/ironic: Bump the "min_command_interval" default value from 5 to 10  https://review.openstack.org/33969116:24
lucasagomesmat128, jroll putting it out to collect thoughts on it ^16:24
*** vishwanathj has quit IRC16:30
*** pcaruana has quit IRC16:31
lucasagomesand I'm calling it a day16:32
lucasagomeshave a great evening all, enjoy the weekend16:32
lucasagomestalk to you all very soon :-)16:32
*** lucasagomes is now known as lucas|afk16:32
sambettso/ lucas|afk16:33
dtantsurhave a nice weekend16:38
*** dtantsur is now known as dtantsur|afk16:38
sambettscya dtantsur|afk16:38
jlvillaldtantsur|afk, Ciao16:38
*** gfhellma has joined #openstack-ironic16:42
openstackgerritVasyl Saienko proposed openstack/ironic: Poll Neutron port status.  https://review.openstack.org/33948916:43
vsaienk0morning jroll, lucasagomes, sambetts, jlvillal and all16:44
jlvillalvsaienk0, Why are you working so late???16:44
jlvillalvsaienk0, You should start your weekend :)16:45
*** daemontool has quit IRC16:45
vsaienk0jlvillal: just finished unit tests for ^, btw, it would be cool if jroll, sambetss, jlvillal have look and let me know if this approach is ok for you :)16:45
vsaienk0I mean https://review.openstack.org/33948916:45
sambettsvsaienk0: Does nova do this for VMs ?16:47
vsaienk0sambetts: what do you mean? set vnic type?16:48
sambettsvsaienk0: no, poll for the ports16:48
vsaienk0yes16:48
vsaienk0sambetts via notifications, but port should change status to ACTIVE, which is not happened at the moment with baremetal ports16:49
sambettscan't we use notifications?16:49
vsaienk0nova uses separate api endpoint for it16:49
vsaienk0I'm not sure that we are ok to have something similar in ironic :)16:50
vsaienk0I'm askink to look to Neutron and Nova changes as well16:50
sambettsvsaienk0: does the API endpoint in nova trickle down to the virt driver/16:50
sambetts?16:50
*** PollyZ_ has quit IRC16:51
*** derekh has quit IRC16:53
vsaienk0sambetts it is a part of virt api16:58
vsaienk0https://github.com/openstack/nova/blob/660ecaee66ccab895b282c2ed45c95c809ad6833/nova/virt/libvirt/driver.py#L4859-L488116:58
*** jaybeale has joined #openstack-ironic16:58
vsaienk0but the question how we will deal with multi-tenancy stuff, when Ironic creates a port16:59
vsaienk0also Ironic doing port-update, so we need to track for port status inside Ironic16:59
devanandaironic should use the ironic service user when creating ports on the cleaning / provisioning network17:00
devanandaand afaik, ironic won't be creating ports on behalf of the tenant (nova does that)17:00
sambettsvsaienk0: right, so I think we should be hooking into these same events, and have after we do the port-update we need to wait for the ports to become active but it should be via notificiation like nova if its availible17:01
sambettsdevananda: correct17:01
vsaienk0devananda: thanks for joining to this conversation. So question how to track Neutron port status17:01
vsaienk0from Ironic, as for me the easiest way is to use polling17:01
vsaienk0sambetts, I doubt that external notifications between services are supported17:02
devanandavsaienk0: there were discussions two or three cycles ago about neutron sending notifications to Nova when port creation / updates were complete17:02
*** trown is now known as trown|lunch17:03
devanandaI do not know the current status of that work - but it is a reasonable (and, IMO, better) architecture17:03
vsaienk0devananda: neutron do it via api call to specific nova endpoint17:03
devanandavsaienk0: gotcha17:03
*** jaybeale has quit IRC17:04
vsaienk0devananda: I don't think it is a god way for Ironic, it is better to use poling from Ironic17:04
devanandavsaienk0: does Nova supply the "call back URL" in some way that we could leverage, if Ironic also had a similar endpoint?17:04
devanandavsaienk0: why?17:04
sambettsvsaienk0: surly we could add a few lines into our virt driver here: https://github.com/openstack/nova/blob/master/nova/virt/ironic/driver.py#L772 which does the same wait_for_instance_events, and the error call_back just tear down the node if our neutroning port updating in Ironic didn't result in an ACTIVE port17:07
vsaienk0sambetts: this solution will work only for tenant ports17:07
vsaienk0it will not work for multi-tenancy provision/cleaning ports17:08
sambettshmm good point17:08
sambettsI think we should find out if we can hook into the same notificiation/event based system if it exists in Neutron17:08
vsaienk0devananda: separate endpoint looks too complicated for me, we receive callback from Neutron, than we need to forward this callback to ironic conductor17:09
sambettsI'm interested how the interaction between nova and neutron is defined17:11
sambettsI was we just had straight forward publish subscribe queues we could listen tooo ...17:12
sambettsI wish*17:12
openstackgerritPavlo Shchelokovskyy proposed openstack/ironic: Update devstack plugin with new auth options  https://review.openstack.org/29731217:16
openstackgerritPavlo Shchelokovskyy proposed openstack/ironic: Migrate to using keystoneauth Sessions  https://review.openstack.org/23698217:16
mat128jroll: all green! https://review.openstack.org/#/c/327891/17:16
sambetts\o/17:17
jrollmat128: woo, approved17:17
mat128:D17:17
vsaienk0https://github.com/openstack/neutron/blob/673cab97a4cd17f8100c3c1e82755b8bea9195e8/neutron/notifiers/nova.py17:17
mat128mission accomplished - landing that before my PTO17:17
mat128:D17:17
mat128thanks17:17
vsaienk0sambetts, devananda: https://review.openstack.org/#/c/327891/17:18
*** amotoki has joined #openstack-ironic17:18
vsaienk0sambetts, devananda: sorry there is the right link https://github.com/openstack/neutron/blob/673cab97a4cd17f8100c3c1e82755b8bea9195e8/neutron/notifiers/nova.py17:18
*** davideagnello has joined #openstack-ironic17:19
*** rcernin has joined #openstack-ironic17:20
*** rcernin has quit IRC17:20
sambettsSo can't we add an Ironic notifier? I think an external node event endpoint in Ironic would be pretty cool actually for things other than this use case17:20
*** jaybeale has joined #openstack-ironic17:22
*** rcernin has joined #openstack-ironic17:24
*** amotoki has quit IRC17:24
mat128sambetts: +1, would help me reverse the dependency on my asset manager :D17:25
*** PollyZ has joined #openstack-ironic17:26
jrollsambetts: you mean emitting notifications on node events?17:28
sambettsjroll: no, an endpoint to receive external events that affect nodes17:29
mat128AHH17:29
mat128sambetts: I understood it the same as jroll17:29
vsaienk0jroll, the related bug: https://launchpad.net/bugs/130467317:29
openstackLaunchpad bug 1304673 in Ironic "Ironic needs to synchronize external events with Neutron" [Medium,In progress] - Assigned to Vasyl Saienko (vsaienko)17:29
mat128sambetts: receiving events that affect nodes could as well just be api calls17:29
jrollsambetts: right, okay17:29
*** sdake_ has joined #openstack-ironic17:30
*** PollyZ has quit IRC17:30
*** gabriel-bezerra is now known as gabriel17:31
*** gabriel is now known as gabriel-bezerra17:31
vsaienk0so for this we need to introduce new state, something like "waiting for network"17:32
sambettsmat128: yeah, I'm thinking a unified endpoint that can receive events, /v1/nodes/<uuid or name>/event and then in our Ironic code we can have wait_for_external_event(event) in places when we need to wait for something externally to happen17:32
mat128hmm17:32
mat128so that your logic doesnt have to be in the caller17:32
mat128seems reasonable17:32
mat128you have to standardize the "events" though17:32
mat128can't you hook onto the existing event notification system?17:33
mat128instance.create.end and friends?17:33
mat128sambetts: https://wiki.openstack.org/wiki/SystemUsageData17:33
sambettsmat128: thats what I would love most, I wish we just had publish subscribe events streams everywhere in OpenStack so you could hook in and your off to the races17:33
*** jpich has quit IRC17:33
*** sdake has quit IRC17:34
mat128sambetts: I guess you could have a ironic-ceilometer-agent or smt17:34
jroll:|17:34
*** sabeen has quit IRC17:35
sambettswe actully need this sort of notification stuff for this too https://github.com/openstack/ironic/blob/master/ironic/dhcp/neutron.py#L14117:39
*** Goneri has quit IRC17:41
*** thiagop is now known as thiagop-afk17:42
rama_yjroll: question on fixing docstring warnings in Ironic?17:43
jrollrama_y: ask away, in a meeting but will try to respond17:44
rama_yjroll: would you like me to submit a patch?17:44
rama_yjroll: that is, if you are not already working on it.17:45
jrollrama_y: that would be helpful, yes, please also turn on warnerrors=True in that patch17:45
jrollI am not17:45
rama_yjroll: ok, sure, thanks. I will post a patch soon.17:46
jrollrama_y: thanks!17:46
* sambetts -> heading off 17:50
sambettshave a good weekend everyone!17:50
mat128cya, have a good one sambetts17:50
*** dprince has quit IRC17:50
*** PollyZ has joined #openstack-ironic17:51
*** amoralej is now known as amoralej|off17:52
*** Sukhdev has joined #openstack-ironic17:53
*** Goneri has joined #openstack-ironic17:57
xavierrgood morning Ironic17:57
*** mbound has quit IRC17:59
*** Sukhdev has quit IRC18:01
harlowjamgould|afk thx for helping get the spec stuff in shape (along with lxsli)18:07
*** Goneri has quit IRC18:09
*** piet has quit IRC18:15
rajinirsambetts: thanks for clarifying the build rechecks. so just to confirm "vendor recheck"  instead of "recheck vendor"18:16
clif_his the execution model for ipa such that its only doing one thing at a time?18:17
clif_has in, races aren't possible, I hope?18:17
*** Goneri has joined #openstack-ironic18:21
*** trown|lunch is now known as trown18:22
openstackgerritMerged openstack/ironic: Add internal_info field to ports and portgroups  https://review.openstack.org/33841718:27
openstackgerritMerged openstack/ironic: Correct reraising of exception  https://review.openstack.org/33943118:28
jrollclif_h: yep: https://github.com/openstack/ironic-python-agent/blob/master/ironic_python_agent/extensions/base.py#L243-L24718:31
jrollas far as commands from the API18:31
clif_hdoesn't the agent driver acquire a lock when sending commands to the agent though?18:32
mat128clif_h: agent will not accept working on more than 1 thing, based on jroll's link18:33
jrollclif_h: yes, but the commands aren't async18:33
jroller, are async18:33
*** electrofelix has quit IRC18:33
jrollclif_h: to be clear, by 'yep' I meant 'one thing at a time is the execution model'18:33
openstackgerritMathieu Mitchell proposed openstack/ironic: Document operators troubleshooting steps  https://review.openstack.org/33605518:34
*** fragatin_ has quit IRC18:34
*** fragatina has joined #openstack-ironic18:35
clif_hgotcha18:37
*** pcaruana has joined #openstack-ironic18:37
xavierrguys, any ideas what to do? http://paste.openstack.org/show/528638/18:39
xavierrit happens when restarting the conductor18:40
*** piet has joined #openstack-ironic18:43
mat128have you run db-sync?18:44
mat128xavierr: ^18:44
xavierrhey mat128, do you mean ironic-dbsync?18:44
mat128xavierr: yes18:45
xavierrnever executed this before18:45
xavierrI'll try18:45
mat128xavierr: well, without context, you're missing a network_interface column in nodes18:45
*** thiagop-afk is now known as thiagop18:50
xavierrmat128, it works! thanks :)18:51
*** _vsaienko has joined #openstack-ironic18:53
*** rama_y has quit IRC18:54
_vsaienkosambetts, jroll, devananda: I will prepare spec about external notifications from Neutron if you do not mind...18:55
*** mbound has joined #openstack-ironic18:59
*** aswadr_ has quit IRC19:01
*** mbound has quit IRC19:04
mat128xavierr: np :)19:11
*** derekh has joined #openstack-ironic19:15
thiagopjlvillal: Thanks for reviewing. I didn't tagged a bug on that change though, mas as clarkb suggested, I did update the commit message to something more apropriate19:15
jlvillalthiagop, thanks19:16
thiagops/mas/but19:16
* thiagop is starting to write in two languages at the same time19:16
* thiagop needs vacation19:16
*** Sukhdev has joined #openstack-ironic19:16
mat128ahem, more vision-sharing to do... https://review.openstack.org/#/c/339543/19:17
*** mat128 is now known as mat128|afk19:18
*** Sukhdev has quit IRC19:18
*** _vsaienko has quit IRC19:22
*** amotoki has joined #openstack-ironic19:22
*** sabeen has joined #openstack-ironic19:24
*** _vsaienko has joined #openstack-ironic19:24
*** amotoki has quit IRC19:27
*** PollyZ has quit IRC19:29
*** Sukhdev has joined #openstack-ironic19:31
*** PollyZ has joined #openstack-ironic19:32
*** piet has quit IRC19:41
*** ChubYann has joined #openstack-ironic19:42
*** Goneri has quit IRC19:45
*** sdake_ has quit IRC19:51
*** garthb has quit IRC20:00
*** garthb has joined #openstack-ironic20:01
*** mtanino has quit IRC20:05
*** mbound has joined #openstack-ironic20:11
*** dprince has joined #openstack-ironic20:14
*** baoli_ has quit IRC20:15
*** mbound has quit IRC20:15
*** gfhellma has quit IRC20:18
*** rama_y has joined #openstack-ironic20:22
*** amotoki has joined #openstack-ironic20:23
mat128|afkvacation time!20:27
mat128|afksee you in 3 weeks :)20:27
*** mat128|afk is now known as mat128|pto20:27
thiagopyooo, have a good time out mat128|pto . Enjoy!20:27
mat128|ptothank you :)20:27
*** amotoki has quit IRC20:27
*** PollyZ has quit IRC20:32
*** trown is now known as trown|outtypewww20:49
*** PollyZ has joined #openstack-ironic20:49
*** rcernin has quit IRC20:52
*** PollyZ has quit IRC20:55
*** daemontool has joined #openstack-ironic21:01
openstackgerritMerged openstack/ironic-python-agent: Replace the ps options when collecting logs  https://review.openstack.org/33894221:02
*** PollyZ has joined #openstack-ironic21:03
*** PollyZ has quit IRC21:08
*** hoangcx has joined #openstack-ironic21:15
*** PollyZ has joined #openstack-ironic21:16
*** PollyZ has quit IRC21:22
*** amotoki has joined #openstack-ironic21:24
*** PollyZ has joined #openstack-ironic21:25
*** amotoki has quit IRC21:28
*** lascii is now known as alaski21:31
*** daemontool has quit IRC21:37
*** wajdi has quit IRC21:40
*** thiagop has quit IRC21:43
*** rbudden has quit IRC21:44
*** rbudden has joined #openstack-ironic21:44
*** dprince has quit IRC21:53
*** ayoung has quit IRC21:55
openstackgerritMerged openstack/ironic-python-agent: Bump CoreOS to 1010.6.0 (last current stable)  https://review.openstack.org/32789122:02
*** thiagop has joined #openstack-ironic22:02
*** thiagop has quit IRC22:03
*** derekh has quit IRC22:10
*** PollyZ has quit IRC22:11
*** superdan is now known as dansmith22:12
*** davidlenwell has quit IRC22:12
*** xhku has quit IRC22:14
*** xhku has joined #openstack-ironic22:14
*** joprovost has quit IRC22:18
*** xhku has quit IRC22:19
*** davidlenwell has joined #openstack-ironic22:22
*** sabeen has quit IRC22:23
*** amotoki has joined #openstack-ironic22:24
*** Sukhdev has quit IRC22:26
*** amotoki has quit IRC22:29
*** Goneri has joined #openstack-ironic22:29
*** PollyZ has joined #openstack-ironic22:31
*** thrash is now known as thrash|wknd22:32
openstackgerritMerged openstack/ironic: Allow to enroll nodes with oneview driver  https://review.openstack.org/32845122:34
openstackgerritRamamani Yeleswarapu proposed openstack/ironic: Fix docstring warnings  https://review.openstack.org/33985222:42
*** rbudden has quit IRC22:44
*** derekh has joined #openstack-ironic22:46
*** rbudden has joined #openstack-ironic22:48
*** PollyZ has quit IRC22:57
*** PollyZ has joined #openstack-ironic23:00
*** PollyZ has quit IRC23:05
*** derekh has quit IRC23:07
*** ppiela has quit IRC23:10
*** PollyZ has joined #openstack-ironic23:14
*** xhku has joined #openstack-ironic23:14
*** xhku_ has joined #openstack-ironic23:16
*** PollyZ has quit IRC23:19
*** xhku has quit IRC23:19
*** xhku_ has quit IRC23:21
*** sabeen has joined #openstack-ironic23:22
*** hoangcx has quit IRC23:23
*** hoangcx has joined #openstack-ironic23:24
*** amotoki has joined #openstack-ironic23:25
*** amotoki has quit IRC23:30
*** ChrisAusten has quit IRC23:34
*** jcoufal has quit IRC23:36
*** xhku has joined #openstack-ironic23:37
*** jaybeale has quit IRC23:38
*** PollyZ has joined #openstack-ironic23:41
*** xhku has quit IRC23:41
*** PollyZ has quit IRC23:46
*** greghaynes has quit IRC23:46
*** Goneri has quit IRC23:54
*** PollyZ has joined #openstack-ironic23:55
*** xhku has joined #openstack-ironic23:57
*** xhku has quit IRC23:58
*** xhku has joined #openstack-ironic23:58
*** greghaynes has joined #openstack-ironic23:58
*** xhku has quit IRC23:59
*** xhku has joined #openstack-ironic23:59
*** ChrisAusten has joined #openstack-ironic23:59
« Thursday, 2016-07-07 Index Saturday, 2016-07-09 »

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!